MicroElectro Mechanical Systems (MEMS) are semiconductor chips that support a top layer of small mechanical devices, such as fluid sensors or mirrors. These devices are built onto chips through growth and etching processes similar to those used to define the topography of an integrated circuit. These processes are capable of creating devices with micron dimensions. The MEMS itself typically packs multiple elements on a single chip.
A MEMS device, specifically a Digital Micromirror Device (DMD), is the basis for Digital Light Processing™ technology. The DMD microchip functions as a fast, reflective digital light switch. The switching is accomplished through the rotation of multiple small mirrors in response to an electric potential. In a mirror's “on” state of rotation, light from a projection source is directed to the pupil of a projection lens and a bright pixel appears on the projection screen. In the “off” state, light is directed out of the pupil and the pixel appears dark. Thus the DMD provides a digital basis for constructing a projected image. Digital Light Processing™ has been employed commercially in televisions, cinemagraphic projection systems, and business-related projectors.
In a typical DMD design, metal, e.g., aluminum, is deposited to form support posts, a hinge, the mirror itself, and structures (such as yokes or landing tips) to contain its rotation. The processes used to define these structures on a DMD (or any other MEMS device) are known in the art and are not the subject of this invention. These processes may include growth of a passivation layer on the mechanical device.
Passivation layers are added to address several problems with device operation. One such problem is static friction (stiction), the static adhesion force between resting bodies in contact (such as two surfaces of a DMD pixel). Another problem is dynamic friction, which arises from the contact of moving elements in the device. Effective passivation layers reduce stiction and friction by reducing the surface energy of the device. For rotating devices (such as the hinge in a DMD), repeated movement and deformation displace molecules and permanently bias the zero state of the rotation. Passivation layers may reduce this hinge memory accumulation by stabilizing certain states of the surface.
Passivation layers are typically formed from surfactants. Effective surfactants are believed to function by forming self-assembled monolayers at the device surface. These monolayers are ordered molecular assemblies formed by the adsorption of a surfactant on a solid surface. Zhu, et. al., “Self-Assembled Monolayer used in Micro-motors,” report the use of such monolayers, formed from an octadecyltrichlorosilane precursor, as a passivation layer for a silicon micromotor. Hornbeck, “Low Surface Energy Passivation Layer for Micromechanical Devices” (U.S. Pat. No. 5,602,671) has described the use of self-assembled monolayers as passivation for MEMS devices including DMDs. Suitable self-assembling carboxylates may be introduced as a vapor under conditions designed to facilitate the growth of self-assembled monolayers, as disclosed by Robbins, “Surface Treatment Material Deposition and Recapture,” (U.S. Pat. No. 6,365,229).
Self-assembled monolayers have been studied outside the device context. Much of the early research in this field concerned the interaction of surfactants with gold surfaces; but work has been published relating to other metals (and metalloids), including silicon and aluminum. Work pertaining to phosphonate/phosphonic acid surfactants includes: Gawalt, et. al, “Self-Assembly and Bonding of Alkanephosphonic Acids on the Native Oxide Surface of Titanium,” Langmuir 2001, 17, 5736-38; Hanson, et. al, “Bonding Self-Assembled, Compact Organophosphonate Monolayers to the Native oxide Surface of Silicon,” J. Am. Chem. Soc. 2003, 125, 16074-80; and Nitowski, G., “Topographic and Surface Chemical Aspects of the Adhesion of Structural Epoxy Resins to Phosphorus Oxo Acid Treated Aluminum Adherents.”
Within the device context, the passivation layer should be stable under the intended operating conditions of the MEMS. While carboxylate surfactants have functioned adequately in commercial DMD products, the resulting monolayers may thermally desorb under foreseeable conditions of operation. Such desorption, and the resulting increase in stiction, friction, and hinge memory accumulation, would adversely impact the operation of the device. It is therefore desirable to form passivation layers from surfactants that bind more tightly with the surface of interest. That is one objective of the present invention. In addition, because of their reduced acidity (as compared to some commonly employed carboxylates), phosphonic acid surfactants may provide compatibility advantages with common packaging materials, such as Kovar.